asgard-ai-platform/med-culture
Use when the user wants to write a culture-and-arts news piece — exhibition coverage, performance reviews, artist interviews, cultural policy analysis, art market reporting, or cultural heritage reporting — from supplied material. Activates culture-beat-specific workflow on top of the general news-reporter workflow: cultural appropriation audit, indigenous knowledge protection, source credibility (institutional PR vs. independent critic), aesthetic judgment marking, and translation/cross-cultural framing discipline. Also triggers on phrases like 'draft an exhibition review', 'write up this gallery opening', 'turn into an album review', '整理展覽新聞', '寫一篇評論', '幫我做藝術家訪談', 'cover this museum exhibition', '草擬一篇文化評論'. Defers general news craft to med-news-reporter; do NOT use for art-marketing copy (use mkt-*) or artist/gallery press releases (use pr-press-release).
npx skills add https://github.com/asgard-ai-platform/skills --skill med-culture
> This skill specializes med-news-reporter for the arts/culture beat. Read med-news-reporter first for the general 6-step workflow (type selection, material audit, fact-check, balance, ethics, literacy). This file adds culture-beat-specific discipline on top: cultural appropriation awareness, indigenous knowledge protection, institutional vs. independent source marking, aesthetic judgment transparency, and cross-cultural translation pitfalls.
Distilled from culture-journalism curricula at Columbia (Arts & Culture Concentration), NYU Journalism, NCCU 文化人類學, NTU 人類學系, 台藝大文化政策所. Covers the main culture-news sub-types: exhibition/museum reporting, performance-arts criticism, artist profile, cultural policy/heritage, and art-market reporting.
IRON LAW: Critique Is Marked Opinion, Not Fact
Aesthetic judgments—"the painting is powerful", "the dancer lacked
precision", "this work pushes boundaries"—are opinions tied to a
critical framework. They must always be (a) marked as judgment, NOT
framed as fact; (b) attributed to a named critic / school of thought
/ institutional voice, with their standpoint disclosed; (c) supported
by reasoning or observable detail, not bare assertion. Factual reporting
(artist X exhibits at venue Y on date Z with medium M) is separable
from aesthetic evaluation. Default LLM behavior conflates them via
descriptive adjectives that smuggle judgment into "fact" ("the brilliant
emerging artist", "the disappointing retrospective"). Override this
explicitly: strip adjectives or attribute them.
Why this is non-obvious: "brilliant" reads like a natural way to
describe an artwork, not an opinion. But it is. The reader deserves
to know whether "brilliant" is (i) a direct quote from the artist's
institution, (ii) an observation by a named art historian with a known
aesthetic stance, or (iii) an invented consensus the writer assumes.
Why this is non-obvious: the aesthetic judgments arrive naturally in prose. A piece on a new exhibition will naturally invite adjectives; defaulting to "powerful" or "striking" silently endorses a reading that may not be universal. Credible critics mark their stance upfront.
Rationalization Table — these justifications DO NOT override the Iron Law:
| Claude might think... | Why it's still a violation |
|---|---|
| "The artist's gallery calls the work 'groundbreaking', so I'll use it" | Still an opinion even if uttered by the institution. Mark it: 「該藝廊稱該作『突破性』」, not 「突破性作品」. |
| "The museum website says 'innovative', it's public info" | Institutional description ≠ fact. Separate: "The museum describes the exhibition as X; art critics debate whether Y." |
| "'Derives from' is just describing influence, not judging" | Influence-attribution is a critical claim. Does the named artist agree? Did they cite this influence themselves, or is it art-historical interpretation? Be explicit. |
| "Critics universally praise this artist, so 'acclaimed' is fact" | "Universally praised" is hyperbole that collapses critical diversity. Even if true, say: "Multiple critics praised..." with names/sources. |
| "The work failed to engage the audience" | Describe observable fact: "audience walked out", "low attendance", not an interpretation. If you infer a judgment, cite whose interpretation. |
| "I'm just translating the original artwork's intent" | Translating artistic intent is interpretation, not fact. Even with the artist's own statement, note: "The artist stated intent is X; critics interpret as Y." |
Trigger conditions:
Input signals:
When NOT to use:
pr-press-release.fin-* or decline.mkt-*.med-entertainment.Read or have already loaded med-news-reporter for: material audit, fact-checking, source-strength tagging, balance principle, media-ethics check, media-literacy self-check. Do not re-implement those steps here. This file specializes Steps 1, 2, 3, and adds a culture-specific Step 7 (Aesthetic Judgment Audit + Cross-Cultural Sensitivity Audit).
| Sub-type | Signals | Sub-template focus |
|----------|---------|--------------------|
| Exhibition/museum | Opening date, venue, artist/curator name, artworks, theme | Curatorial intent + artwork description separate from judgment; source marking (PR vs. independent review) |
| Performance review | Venue, troupe, performer, date, genre (劇場/音樂/舞蹈), attendance | Aesthetic judgment tied to named critic or review publication; descriptive detail before evaluation |
| Artist profile | Artist bio, background, major works, recent exhibition/residency | Career trajectory via public record; quotes attributed; avoid "rising star" unless cited |
| Cultural policy/heritage | 文化部政策、文資法、世界遺產、古蹟保存、文化補助機制 | Policy text + stakeholder impact + indigenous/minority voice included; legal/administrative accuracy |
| Art market | 拍賣、藝術基金、畫廊銷售、藝術家身價、收藏趨勢 | Price ≠ artistic quality (separate fact from market signal); conflict-of-interest disclosure (auction house estimator, dealer stake) |
If the material spans sub-types (e.g., an artist's first major museum show), classify by the *primary news driver*.
Every cultural institution quoted must be tagged for PR affinity at first mention:
Source tier tagging (extends med-news-reporter's tiering with culture-specific tiers):
| Tier | Examples | Treatment |
|------|----------|-----------|
| Public record / official document | 文化部施政報告、世界遺產名單、古蹟登錄公告、展覽官方新聞稿 | Direct citation; note source is institutional. |
| Curatorial/artist statement (on record) | 策展人訪談、藝術家自述、美術館展覽說明頁面 | Direct quote with role tag; note: this is institution's framing, not independent view. |
| Independent critic/reviewer | 藝評人刊登於獨立媒體之評論、學者論文、展評 | Direct quote; include publication/credentials to let reader calibrate standpoint. |
| Collector / dealer | 拍賣公司、畫廊、私人收藏家 | Quote with interest disclosure: "Sotheby's estimated [price]" signals auction-house stake in high valuation. |
| Audience/community | Visitor comment, indigenous/minority community perspective on cultural heritage | Include voice especially when discussing cultural appropriation, indigenous knowledge, or community impact. |
Beyond med-news-reporter's general ethics check, add:
references/ethics_and_law.md for fair-use thresholds.Before output, sweep the draft for:
Use the med-news-reporter base format, with these culture additions to the meta footer:
[Headline / sub-headline / body paragraphs per med-news-reporter]
---
**稿件類型**: 即時展覽新聞 / 演出評論 / 藝術家專訪 / 文化政策追蹤 / 藝術市場分析
**字數**: approx. XXX
**消息來源層級**: 公開記錄 N / 機構 (PR) N / 獨立評論 N / 藝術家/策展人 N
**文化敏感度稽核**:
- 機構聲音標記: ✅ / ⚠️ (列出未標記的機構引述)
- 獨立評論或平衡: ✅ / ⚠️ (機構PR vs. 獨立評論的比例)
- 審美判斷標記: ✅ / ⚠️ (列出未標記的評價形容詞)
- 文化挪用檢查: ✅ / N/A / ⚠️
- 原住民族聲音: ✅ / N/A / ⚠️ (涉及時)
- 翻譯 / 跨文化框架: ✅ / N/A / ⚠️
- 價格 ≠ 價值分離: ✅ / N/A / ⚠️
**待查證事項**: ...
**倫理 / 識讀檢核摘要**: 〔交給 med-news-reporter 的 Step 4-5 footer〕
See examples/sample_input.md and examples/sample_output.md for a full worked exhibition-reporting scenario.
| File | Purpose | When to read |
|------|---------|--------------|
| references/sources_and_beats.md | 文化線消息來源(文化部、各美術館、評論刊物、國際資源) | Step 2 source vetting |
| references/glossary.md | 文化、藝術、文資領域術語對照與定義 | When unfamiliar terminology appears |
| references/ethics_and_law.md | 著作權法合理使用、文化資產保存法、原住民族傳統智慧創作保護、肖像權 | Step 3 risk check |
| references/cultural_sensitivity.md | 文化挪用辨識、原住民族報導守則、宗教/儀式報導框架、跨文化翻譯陷阱 | Step 4-5 sensitivity audits |
| references/critique_writing.md | 描述 vs 評價之分界、評論的事實基礎、價格≠價值、市場趨勢≠藝術品質 | Step 4 aesthetic judgment audit |
Related skills:
med-news-reporter — general news workflow (this skill specializes it)med-entertainment — for film/TV/streaming-specific coverage (separate beat focus)hum-source-criticism — source vetting frameworkshum-rhetoric — deeper analysis of persuasion structures (useful for critique analysis)grad-ethnography — ethnographic / cultural-context research methodologyTake asgard-ai-platform/med-culture from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.